Ординатура / Офтальмология / Английские материалы / Retinal Development_Sernagor, Eglen, Harris, Wong_2006

up-regulated genes (Opn1sw, Gnat2), at least one subset of these genes are expressed either in mitotic progenitors (Fgf15), M¨uller glia (Fabp7, ApoE, Gdc), or amacrine cells (Dfy). As is the case with the Brn3b mutant animals, disruption of development of a specific cell type clearly produces a variety of cell non-autonomous effects on gene expression. The mechanism by which this occurs, and the consequences for the development of other cells in the retina, has yet to be explored.

One published study has used high-throughput in situ hybridization as an initial screen to identify genes that show dynamic cellular expression patterns during retinal development. In this study (Thut et al., 2001), 1035 individual IMAGE consortium cDNAs were examined against a series of retinal sections from E12, E13, E14, E16, E18 and P2 mice. Though these cDNAs were selected on the basis of known or putative roles in regulating development, only 17 genes tested showed clear cell-specific expression in the developing neuroretina, 5 of which had been previously reported to be expressed during retinal development. Fgf15 was confirmed as a prominent marker of retinal progenitors, while Ptmb10 and thyroid hormone receptor alpha were strongly expressed in the inner neuroblastic layer. Six additional genes were selectively expressed at the ciliary margin, including Tgfbli14 and Ptmb4, and it was shown that expression of these two genes could be induced in E10 to E13 neuroretina by direct contact with cocultured chick lens.

So far, only one study has combined both expression profiling with large-scale in situ hybridization analysis of genes that were observed to show differential expression in developing retina. In this final study, SAGE libraries were constructed from mouse retinas at two-day intervals from E12.5 to P6.5, with libraries also being made from P10.5 and P50 retina (Blackshaw et al., 2004). A total of 1051 genes that showed dynamic expression during development by either visual inspection or cluster analysis were examined by in situ hybridization of a panel of retinal sections that recapitulated the time course of the SAGE libraries. A molecular atlas of gene expression in the developing and mature retina was thereby constructed, along with a taxonomic classification of developmental gene expression patterns.

Genes were identified that label both temporal and spatial subsets of mitotic progenitor cells, thus identifying potential molecular mediators of changes in the developmental competence of progenitors and demonstrating that there is considerable heterogeneity among progenitors. For each developing and mature major retinal cell type, genes selectively expressed in that cell type were identified. The gene expression profiles of retinal M¨uller glia and mitotic progenitor cells were found to be highly similar, suggesting that M¨uller glia might serve to produce multiple retinal cell types under the right conditions. Strong and transient expression of many metabolic enzymes was observed in immature M¨uller glia, though the significance of this finding is unclear. In addition, multiple transcripts that did not appear to encode open reading frames (ORFs) that were evolutionarily conserved or <100 amino acids in length (‘non-coding RNAs’) were found to be dynamically and specifically expressed in developing and mature retinal cell types. Finally, many photoreceptor-enriched genes that mapped to chromosomal intervals containing uncloned retinal disease genes were identified.

16.4 Comparing the results of the mouse studies

Four studies have surveyed a large fraction of the mouse retinal transcriptome at similar time-points, and thus fruitful comparisons can be made. Serial analysis of gene expression tags are detected that match the majority of the 800-odd transcripts enriched in 4N cells of the developing retina, with 125 genes being present at >0.1% of tags in at least one SAGE library (see Supplementary Table 16.2). The temporal expression profile of 84% of these genes was generally falling over time, with 13% basically unchanging and only 3% rising through development – just as would be expected for progenitor-enriched genes. Twenty-two of these genes were tested by in situ hybridization in the SAGE study, and the great majority were found to be progenitor-enriched, with some showing broad expression throughout mitosis and some showing restricted expression in subsets of cells. When only genes for which probes are present on the array and genes sufficiently abundant to be detected by SAGE are considered, agreement between the two data sets is quite good. A caveat must be made that it is usually the more abundant and dynamically expressed genes in any primary data set that get examined by in situ hybridization, so the very high truepositive rate seen here probably represents an overestimate when considering the entire data set.

The time course of gene expression in the postnatal retina as measured by both SAGE and Affymetrix chips could also be compared. Eight hundred and seventy-five of the 5249 genes detected by the arrays in the postnatal retina were also present at >0.1% of tags in at least one SAGE library (see Supplementary Table 16.3). Here again, there is reasonable correlation between the strongly expressed genes in the two data sets, particularly for genes that show steadily increasing expression through development. Many of these turn out to be photoreceptor-enriched genes, although these also include genes selectively expressed in every other major types of retinal cell. Combining these results from the array data set obtained from Nrl−/− animals reveals a core set of 41 genes that are very likely to be rodenriched, even in the absence of cellular expression data, on the basis of showing both late onset of expression by both SAGE and Affymetrix chips as well as reduced expression in Nrl−/− animals. This includes a total of ten genes whose cellular expression pattern in the retina has not yet been previously analysed (see Supplementary Table 16.3). As a general rule, however, the more highly expressed a gene, the more likely the SAGE and array data are to correlate with one another, and low-abundance transcripts often show little or no correlation between the two data sets.

16.5 What have genomics studies told us about mechanisms of retina development?

So far, these genomic studies have left us at a tantalizing but somewhat frustrating point. They have given us an extensive but incomplete parts list for the assembly of a mature retina

– potentially the starting point of 1000 projects – and have given us a vastly expanded list of cell-specific markers. Table 16.3 contains a partial list for the mouse retina. However, these

Table 16.3. Cell-specific markers identified in mouse retina using genomic approaches

studies have not yet given us much coherent data about mechanisms of retinal development. The fact that no method of expression profiling gives a fully comprehensive picture of gene expression has limited our ability to extract patterns from the data. None of these studies has yet moved to functional analysis of any of the developmentally dynamic genes identified in these studies, so we still have little clue what mechanistic role these differentially expressed genes play in retinal development.

In studies where expression profiling was combined with large-scale in situ hybridization of differentially regulated genes, the sheer number of the cellular expression patterns examined has allowed some general conclusions to be drawn. These studies have also reinforced some previously anticipated details regarding retinal development. Dynamic waves of expression of selected genes are observed in mitotic progenitors, as would be predicted on the basis of the changing developmental competence of retinal progenitors (Cepko et al., 1996). A great diversity of expression patterns is observed for amacrine-enriched genes, as might be expected from the considerable morphological diversity of amacrine cells (MacNeil and Masland, 1998).

These studies have also identified some general, unanticipated themes in retinal development. These include the considerable heterogeneity of mitotic progenitors, the similarity of mitotic progenitors and M¨uller glia, the dramatically elevated expression of various metabolic enzymes in newly formed glia, and the prominent and dynamic expression of mRNA-like molecules that do not encode proteins in developing retina. Genomic studies have also yielded a potential wealth of clinically important data. Nearly half of all cloned Mendelian photoreceptor dystrophy genes are highly and selectively expressed in either developing or mature photoreceptors (Blackshaw et al., 2001; Katsanis et al., 2002; Pacione et al., 2003; Yu et al., 2004b). As a result, the genomic studies aimed at identifying photoreceptor-enriched genes have uncovered a wealth of potential candidate genes for inherited photoreceptor dystrophies that have not yet been cloned. The results of one SAGE-based study were used to pinpoint the photoreceptor-enriched IMPDH1 gene as being mutated in RP10 patients (Bowne et al., 2002). Other genes that were later implicated in retinal disease or photoreceptor survival, including RDH12 (Haeseleer et al., 2002; Janecke et al., 2004), BBS5 (Li et al., 2004), and rod-derived cone survival factor (Leveillard et al., 2004), were also prospectively identified as photoreceptor-enriched genes using genomics approaches (although the approaches used to clone these genes did not directly use this genomic data). The catalogue of disease genes identified using genomic approaches is likely to grow considerably in the years ahead.

It remains the case, however, that for most genomic expression studies, the cellular expression patterns of differentially expressed genes have not been examined. Without this data, it is generally not yet possible to draw broader mechanistic conclusions about retinal development from these genomic data. The situation is somewhat clearer for studies that examine mutant animals, such as those performed on mice deficient for Nrl, Rb1, and Brn3b. In these cases, clear sets of cell-specific genes are disregulated. Given that the mutated genes are transcription factors, it’s likely that at least some of the differentially expressed genes will indeed represent direct targets of those factors. Though some progress

has been made on this topic, particularly in the demonstration that the BMP/Smad pathway modulates expression of rod-specific genes (Yu et al., 2004a), on the whole this awaits further experimental investigation.

16.6 Future directions

There is clearly an essential need to broaden and deepen the body of genomic data on retinal development, and to systematically characterize the cellular expression patterns of genes already identified in previous studies. However, beyond descriptive anatomical studies lies the promise of functional genomics. Every one of the studies described here has generated many genes that can be fruitfully investigated by conventional reverse genetic and biochemical approaches. Nonetheless, the availability of collections of both expressible full-length cDNAs and small interfering RNAs that cover a large fraction of the Drosophila and mouse genome (Berns et al., 2004; Boutros et al., 2004; Paddison et al., 2004; Zheng et al., 2004), along with means of delivering them to the developing retina in vivo through electroporation or viral transduction, implies that medium-throughput functional examination of large numbers of genes identified in these studies may soon be practical.

However, perhaps the main outstanding challenge in developmental genomics is not insufficient data, or a lack of means to examine the function of genes of interest, but rather the poor accessibility of most of the data to the research community. Several factors have limited the ability of the community to access and formulate hypotheses from the data produced by these studies. First, direct comparison of data obtained by different studies is made difficult by the lack a common identifier for individual genes. Second, the great majority of expression data in these studies is usually contained in the supplementary data of publications, and is neither deposited in publicly available databases nor visible to search engines such as PubMed or Google. Expression data for a given gene can thus only be found by combing each paper for the specific reference.

An agreement on the part of the research community to adopt a common intuitive vocabulary for gene identification, and to deposit published expression data in public databases, would go a long way to addressing these problems and greatly increase the usefulness of this data to the broader research community.

References

Berns, K., Hijmans, E. M., Mullenders, J. et al. (2004). A large-scale RNAi screen in human cells identifies new components of the p53 pathway. Nature, 428, 431–7.

Blackshaw, S., Fraioli, R. E., Furukawa, T. and Cepko, C. L. (2001). Comprehensive analysis of photoreceptor gene expression and the identification of candidate retinal disease genes. Cell, 107, 579–89.

Blackshaw, S., Kuo, W. P., Park, P. J. et al. (2003). MicroSAGE is highly representative and reproducible but reveals major differences in gene expression among samples obtained from similar tissues. Genome Biol., 4, R17.

Blackshaw, S., Harpavat, S., Trimarchi, J. et al. (2004). Genomic analysis of mouse retinal development. PLoS Biol, 2, E247.

Boutros, M., Kiger, A. A., Armknecht, S. et al. (2004). Genome-wide RNAi analysis of growth and viability in Drosophila cells. Science, 303, 832–5.

Bowne, S. J., Sullivan, L. S., Blanton, S. H. et al. (2002). Mutations in the inosine monophosphate dehydrogenase 1 gene (IMPDH1) cause the RP10 form of autosomal dominant retinitis pigmentosa. Hum. Mol. Genet., 11, 559–68.

Brenner, S., Johnson, M., Bridgham, J. et al. (2000). Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays. Nat. Biotechnol., 18, 630–4.

Cepko, C. L., Austin, C. P., Yang, X., Alexiades, M. and Ezzeddine, D. (1996). Cell fate determination in the vertebrate retina. Proc. Natl. Acad. Sci. U. S. A., 93, 589–95.

Cheng, J., Kapranov, P., Drenkow, J. et al. (2005). Transcriptional maps of 10 human chromosomes at 5-nucleotide resolution. Science, 308, 1149–54.

Chowers, I., Liu, D., Farkas, R. H. et al. (2003). Gene expression variation in the adult human retina. Hum. Mol. Genet., 12, 2881–93.

Diaz, E., Yang, Y. H., Ferreira, T. et al. (2003). Analysis of gene expression in the developing mouse retina. Proc. Natl. Acad. Sci. U. S. A., 100, 5491–6.

Dorrell, M. I., Aguilar, E., Weber, C. and Friedlander, M. (2004). Global gene expression analysis of the developing postnatal mouse retina. Invest. Ophthalmol. Vis. Sci., 45, 1009–19.

Evans, S. J., Datson, N. A., Kabbaj, M. (2002). Evaluation of Affymetrix Gene Chip sensitivity in rat hippocampal tissue using SAGE analysis. Serial Analysis of Gene Expression. Eur. J. Neurosci., 16, 409–13.

Gawantka, V., Pollet, N., Delius, H. et al. (1998). Gene expression screening in Xenopus identifies molecular pathways, predicts gene function and provides a global view of embryonic patterning. Mech. Dev., 77, 95–141.

Gong, S., Zheng, C., Doughty, M. L. et al. (2003). A gene expression atlas of the central nervous system based on bacterial artificial chromosomes. Nature, 425, 917–25.

Gustincich, S., Contini, M., Gariboldi, M. et al. (2004). Gene discovery in genetically labeled single dopaminergic neurons of the retina. Proc. Natl. Acad. Sci. U. S. A., 101, 5069–74.

Hackam, A. S., Bradford, R. L., Bakhru, R. N. et al. (2003). Gene discovery in the embryonic chick retina. Mol. Vis., 9, 262–76.

Haeseleer, F., Jang, G. F., Imanishi, Y. et al. (2002). Dual-substrate specificity short chain retinol dehydrogenases from the vertebrate retina. J. Biol. Chem., 277, 45 537–46.

Irizarry, R. A., Warren, D., Spencer, F. et al. (2005). Multiple-laboratory comparison of microarray platforms. Nat. Methods, 2, 345–50.

Janecke, A. R., Thompson, D. A., Utermann, G. et al. (2004). Mutations in RDH12 encoding a photoreceptor cell retinol dehydrogenase cause childhood-onset severe retinal dystrophy. Nat. Genet., 36, 850–4.

Katsanis, N., Worley, K. C., Gonzalez, G., Ansley, S. J. and Lupski, J. R. (2002). A computational/functional genomics approach for the enrichment of the retinal transcriptome and the identification of positional candidate retinopathy genes. Proc. Natl. Acad. Sci. U. S. A., 99, 14 326–31.

Kennan, A., Aherne, A., Palfi, A. et al. (2002). Identification of an IMPDH1 mutation in autosomal dominant retinitis pigmentosa (RP10) revealed following comparative microarray analysis of transcripts derived from retinae of wild-type and Rho(−/−) mice. Hum. Mol. Genet., 11, 547–57.

Leveillard, T., Mohand-Said, S., Lorentz, O. et al. (2004). Identification and characterization of rod-derived cone viability factor. Nat. Genet., 36, 755–9.

Li, J. B., Gerdes, J. M., Haycraft, C. J. et al. (2004). Comparative genomics identifies a flagellar and basal body proteome that includes the BBS5 human disease gene. Cell, 117, 541–52.

Livesey, F. J., Furukawa, T., Steffen, M. A., Church, G. M. and Cepko, C. L. (2000). Microarray analysis of the transcriptional network controlled by the photoreceptor homeobox gene Crx. Curr. Biol., 10, 301–10.

Livesey, F. J., Young, T. L. and Cepko, C. L. (2004). An analysis of the gene expression program of mammalian neural progenitor cells. Proc. Natl. Acad. Sci. U. S. A., 101, 1374–9.

MacNeil, M. A. and Masland, R. H. (1998). Extreme diversity among amacrine cells: implications for function. Neuron, 20, 971–82.

Margulies, M., Egholm, M., Altman, W. E. et al. (2005). Genome sequencing in microfabricated high-density picolitre reactors. Nature, 437, 376–80.

Mears, A. J., Kondo, M. and Swain, P. K. (2001). Nrl is required for rod photoreceptor development. Nat. Genet., 29, 447–52.

Mecham, B. H., Klus, G. T., Strovel, J. et al. (2004). Sequence-matched probes produce increased cross-platform consistency and more reproducible biological results in microarray-based gene expression measurements. Nucleic Acids Res., 32, e74.

Mu, X., Beremand, P. D., Zhao, S. et al. (2004). Discrete gene sets depend on POU domain transcription factor Brn3b/Brn-3.2/POU4f2 for their expression in the mouse embryonic retina. Development, 131, 1197–210.

Neidhardt, L., Gasca, S., Wertz, K. et al. (2000). Large-scale screen for genes controlling mammalian embryogenesis, using high-throughput gene expression analysis in mouse embryos. Mech. Dev., 98, 77–94.

Pacione, L. R., Szego, M. J., Ikeda, S., Nishina, P. M. and McInnes, R. R. (2003). Progress toward understanding the genetic and biochemical mechanisms of inherited photoreceptor degenerations. Annu. Rev. Neurosci., 26, 657–700.

Paddison, P. J., Silva, J. M., Conklin, D. S. et al. (2004). A resource for large-scale RNA-interference-based screens in mammals. Nature, 428, 427–31.

Pritchard, C. C., Hsu, L., Delrow, J. and Nelson, P. S. (2001). Project normal: defining normal variance in mouse gene expression. Proc. Natl. Acad. Sci. U. S. A., 98, 13 266–71.

Shendure, J., Porreca, G. J., Reppas, N. B. et al. (2005). Accurate multiplex polony sequencing of an evolved bacterial genome. Science, 309(5741), 1728–32.

Thut, C. J., Rountree, R. B., Hwa, M. and Kingsley, D. M. (2001). A large-scale in situ screen provides molecular evidence for the induction of eye anterior segment structures by the developing lens. Dev. Biol., 231, 63–76.

Tietjen, I., Rihel, J. M., Cao, Y. et al. (2003). Single-cell transcriptional analysis of neuronal progenitors. Neuron, 38, 161–75.

Tomancak, P., Beaton, A., Weiszmann, R. et al. (2002). Systematic determination of patterns of gene expression during Drosophila embryogenesis. Genome Biol, 3, RESEARCH0088.

Velculescu, V. E., Zhang, L., Vogelstein, B. and Kinzler, K. W. (1995). Serial analysis of gene expression. Science, 270, 484–7.

Yoshida, S., Yashar, B. M., Hiriyanna, S. and Swaroop, A. (2002). Microarray analysis of gene expression in the aging human retina. Invest. Ophthalmol. Vis. Sci., 43, 2554–60.

Yoshida, S., Mears, A. J., Friedman, J. S. et al. (2004). Expression profiling of the developing and mature Nrl−/− mouse retina: identification of retinal disease candidates and transcriptional regulatory targets of Nrl. Hum. Mol. Genet., 13, 1487–503.

Yu, J., He, S., Friedman, J. S. et al. (2004a). Altered expression of genes of the Bmp/Smad and Wnt/calcium signaling pathways in the cone-only Nrl−/− mouse retina, revealed by gene profiling using custom cDNA microarrays. J. Biol. Chem., 279, 42 211–20.

Yu, J., Mears, A. J., Yoshida, S. et al. (2004b). From disease genes to cellular pathways: a progress report. Novartis Found. Symp., 255, 147–60; discussion 160–4, 177–8.

Zhang, J., Gray, J., Wu, L. et al. (2004). Rb regulates proliferation and rod photoreceptor development in the mouse retina. Nat. Genet., 36, 351–60.

Zheng, L., Liu, J., Batalov, S. et al. (2004). An approach to genomewide screens of expressed small interfering RNAs in mammalian cells. Proc. Natl. Acad. Sci. U. S. A., 101, 135–40.

and the appearance of an interesting phenotype is sought in subsequent generations. The mutated gene leading to the phenotype is isolated, cloned and sequenced. Not only can the function of the mutated gene be elucidated by this method, but also fundamental cellular or behavioural processes can be studied in the absence of the specific gene product. Following the pioneering work at the University of Oregon, two laboratories developed methods for efficient and large-scale chemical mutagenesis of zebrafish for the expressed purpose of identifying recessive mutations affecting embryonic development (Mullins and Nusslein-Volhard, 1993; Driever et al., 1994; Solnica-Krezel et al., 1994). Both screens used the alkylating agent N-ethyl-N-nitrosourea (ENU) to induce point mutations in zebrafish spermatagonia. The effectiveness of ENU mutagenesis typically generates more than one mutant phenotype per genome. Recessive mutations are then recovered in a traditional third-generation screen. The high rate of mutagenesis combined with morphological analysis, enabled the isolation of thousands of mutations affecting hundreds of loci essential to development of the vertebrate embryo, including the eye and visual system (Brockerhoff et al., 1995; Baier et al., 1996; Driever et al., 1996; Haffter et al., 1996; Malicki et al., 1996). These methods subsequently have been adopted for several small-scale highly focused screens to identify mutations specific to affecting the development and function of the retina (Fadool, et al., 1997; Li and Dowling, 1997; Vihtelic, et al., 2001; Perkins et al., 2002).

Another major advance in zebrafish forward genetic screens occurred with the application of a pseudotype retrovirus vector for insertional mutagenesis (Lin et al., 1994; Gaiano et al., 1996; Golling et al., 2002; Amsterdam et al., 2004). First developed as a vector for gene therapy and genetic studies, the engineered virus can infect a wide range of organisms and efficiently integrate into the genome. In zebrafish, transformation rates are approaching 100%, with most founders transmitting on average 10 proviral inserts to progeny. One in 80 inserts results in an embryonic lethal mutation, and in a large-scale screen hundreds of insertional mutations were recovered over a several year period, although only a fraction of these resulted in specific developmental phenotypes. It is estimated that the 315 saved mutants reflect 25% of the genes essential for the development of many different embryonic structures and organs including the eye (Allende et al., 1996; Becker et al., 1998; Amsterdam et al., 2004; Gross et al., 2005). Comparisons to other species, namely Saccharomyces cervesiae and Caernorhabditis elegans, revealed that 77% and 72% respectively of the essential fish genes are evolutionarily essential in the other species. One clear advantage of insertional mutagenesis is that the proviral insert acts as a molecular tag that facilitates the rapid cloning of the mutated gene as compared with the laborious effort required for positional cloning of ENU-induced mutations (Gaiano et al., 1996). However, the use of the retrovirus techniques in modestly sized screens or to isolate specific phenotypes may be limited.

Although embryonic stem cells and targeted mutagenesis have not been developed for zebrafish, alternative reverse genetic and gene knock-down methods in zebrafish have enabled the analysis of phenotypes of known genes. Targeting induced local lesions in genomes (TILLING), a method originally developed for plant mutagenesis screens, has been used successfully to identify genetic lesions in specific genes of interest (McCallum